Journal
IEEE TRANSACTIONS ON INTELLIGENT VEHICLES
Volume 3, Issue 4, Pages 585-594Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIV.2018.2874530
Keywords
Usage-based-insurance; driving behaviors; auto-mobile insurance
Categories
Funding
- Ministry of Science and Technology, Taiwan [MOST 106-2221-E-155-066, MOST 106-2218-E-155-007, MOST 105-2218-E-155-003, MOST 105-2218-E-155-010]
Ask authors/readers for more resources
Usage-based-insurance is an emerging automobile insurance service in which the driver premium is set individually for each policyholder. A personalized automobile insurance mechanism presents challenges that differ from those presented by the general driver assistance applications that analyze driver behaviors. In this paper, a novel framework based on boosted multiplekernel learning is proposed to reflect the driving risk level of each driver for automobile usage-based-insurance. In the proposed framework, a set of kernels is specified to represent the inherent characteristics of vehicle-oriented, driver-oriented, and lane-oriented attributes. These multiple kernels are carefully integrated using the AdaBoost technique to realize particular collaborative features for driving risk assessment. Experimental results obtained using a lab-recorded driving data set under real-world conditions reveal that the proposed framework exhibits impressive accuracy and robustness in terms of different driving-risk levels.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available